Revised Version of Review Article for Atmospheric Research Satellite Contributions to the Quantitative Characterization of Biomass Burning for Climate Modeling Charles Ichoku, Ralph Kahn, Mian Chin Lab. for Atmospheres, NASA Goddard Space Flight Center, Greenbelt, Maryland, USA Corresponding Author: Dr. Charles Ichoku Climate & Radiation Branch, Code 613.2 NASA Goddard Space Flight Center Greenbelt, MD 20771, USA Phone : +1-301-614-6212 Fax : +1-301-614-6307 or +1-301-614-6420 Email : [email protected] Abstract Characterization of biomass burning from space has been the subject of an extensive body of literature published over the last few decades. Given the importance of this topic, we review how satellite observations contribute toward improving the representation of biomass burning quantitatively in climate and air-quality modeling and assessment. Satellite observations related to biomass burning may be classified into five broad categories: (i) active fire location and energy release, (ii) burned areas and burn severity, (iii) smoke plume physical disposition, (iv) aerosol distribution and particle properties, and (v) trace gas concentrations. Each of these categories involves multiple parameters used in characterizing specific aspects of the biomass-burning phenomenon. Some of the parameters are merely qualitative, whereas others are quantitative, although all are essential for improving the scientific understanding of the overall distribution (both spatial and temporal) and impacts of biomass burning. Some of the qualitative satellite datasets, such as fire locations, aerosol index, and gas estimates have fairly long-term records. They date back as far as the 1970s, following the launches of the DMSP, Landsat, NOAA, and Nimbus series of earth observation satellites. Although there were additional satellite launches in the 1980s and 1990s, space-based retrieval of quantitative biomass burning data products began in earnest following the launch of Terra in December 1999. Starting in 2000, fire radiative power, aerosol optical thickness and particle properties over land, smoke plume injection height and profile, and essential trace gas concentrations at improved resolutions became available. The 2000s also saw a large list of other new satellite launches, including Aqua, Aura, Envisat, Parasol, and CALIPSO, carrying a host of sophisticated instruments providing high quality measurements of parameters related to biomass burning and other phenomena. These improved data products have enabled significant progress in the study of biomass burning from space. However, appreciable uncertainty remains in many of the measurements that still needs to be addressed. Nevertheless, climate and other atmospheric models are 1 making significant adjustments to take advantage of quantitative satellite measurements in studying biomass burning activity, emissions, and impacts. New research directions should include not only improvements in satellite retrievals and modeling accuracies, but also increased synergy between them, such that satellite measurements can be directly input into models without requiring elaborate interpretation. 1 Introduction Biomass burning is a widespread phenomenon affecting most vegetated parts of the world seasonally, either in the form of wildfires ignited by accident or by natural causes such as lightning, or prescribed fires used for agricultural, ecological control or other similar purposes (e.g. Andreae, 1991; Carmona-Moreno et al., 2005; e.g. Morton et al., 2008). Collectively, such large fires are among the fastest agents of terrestrial ecosystem change. Whereas prescribed fires are mainly intended for beneficial purposes, wildfires can have both direct and indirect adverse effects on human life and property, the environment (degradation, soil destabilization, and desertification), and water resources (soil moisture depletion and water pollution). They can also alter air circulation (convection and entrainment of smoke and subsequently soil particles), cloud formation and dissipation, and can effect Earth surface albedo changes. All of these have the potential to exert significant climate impacts that can trigger additional adverse responses and feedbacks (e.g. Cochrane, 2003; Shakesby and Doerr, 2006; Randerson et al., 2006; Bowman et al., 2009). Smoke emitted by fires is composed of aerosol particulate matter (PM) and numerous trace gases, including carbon monoxide (CO), carbon dioxide (CO2), methane (CH4), non-methane hydrocarbons, halogenated compounds, nitrogen oxides (NOx), and volatile organic compounds (VOCs), most of which are pollutants and contribute to the formation of new pollutants, such as tropospheric ozone (O3) and secondary aerosols. These PM and trace gases can have significant impacts, not only on air quality and health, but some (e.g. PM and the greenhouse gases, CO2 and CH4) also affect climate, with potential feedback on air quality. For instance, smoke PM can influence precipitation processes resulting in delayed, suppressed, or invigorated rainfall (e.g. Rosenfeld et al., 1999; Andreae et al., 2004; Koren et al., 2004), change cloud albedo, and scatter and absorb solar radiation, affecting atmospheric warming or cooling, and contributing to climate change. Conversely, expected climatic-changes, such as more severe drought conditions in some regions, are likely to result in more frequent and possibly more severe wildfire events. These complex and multi-faceted impacts of biomass burning can be better characterized and understood only through accurate, quantitative assessment of the spatial and temporal patterns in fuel consumption, heat budgets, and emissions (e.g. Radke et al. 2000, Clements et al. 2007, Kremens et al. 2010). In particular, accurate estimation of smoke emission source strength from active fires is essential for modeling the smoke particulate and gaseous species fluxes, transport, atmospheric interactions, and impacts on air quality and climate. For several decades, researchers have made efforts to estimate burned biomass and smoke emissions from ground-based and in situ measurements, but the spatial and temporal coverage is severely limited (e.g. Crutzen and Andreae, 1990; Andreae and Merlet, 2001; Reid et al., 2005a,b). The rapid growth of satellite measurement capability 2 within the last couple of decades has provided the potential to overcome these space and time limitations by covering the entire globe frequently, over long periods of time. Henceforth, satellite data represent the primary source of information for mapping biomass-burning activity and evaluating smoke emissions at regional-to-global scales (e.g. Schultz et al., 2002; Freitas et al., 2005; Davies et al., 2009; Ichoku et al., 2008a; Kahn et al., 2008; Reid et al., 2009; Val Martin et al., 2010; van der Werf et al., 2010). However, satellite remote-sensing methods are faced with new challenges as they attempt to use instantaneous observational snapshots to address continuous and highly variable processes such as fires and their emissions. The result is that, although satellite can cover more ground, uncertainties in quantifying emissions still remain, and can in some respects be even greater, compared to the ground-based methods. The aim of this review is to assess the contributions satellite remote sensing make to the quantitative characterization of biomass burning for air quality and climate modeling applications. As biomass burning is a vastly interdisciplinary subject whose many aspects have been explored for decades by scholars from different perspectives, including by laboratory and field experimentation, modeling, ground-based, airborne, and satellite approaches, this paper cannot possibly provide an exhaustive review of the subject matter. Rather, we focus the discussion on the contributions from satellite, by way of cataloguing the significant satellite measurements that are directly relevant to the study of biomass burning and its impacts. We examine the current uncertainty levels of some of these satellite products, discuss their current or potential uses, and address some of the limitations and gaps that still exist in these satellite products. We also summarize their existing or potential synergy with modeling that does or could help improve scientific understanding of the biomass burning phenomenon and its impacts, quantitatively, at regional to global scales. Scientific understanding of biomass burning and its impacts requires a fundamental knowledge of three important aspects: (1) the types and spatio-temporal distributions of biomass burning events, (2) the different physical components of a biomass burning process, and (3) the products of biomass burning, their properties, and their trajectories. These three aspects are discussed in more detail in the subsections that follow, within this introduction. Section 2 describes the satellite observational constraints in measuring basic parameters related to biomass burning, including brief highlights of their current uncertainty levels and some of their relevant applications. Section 3 examines the three aspects of modeling that are most frequently used in biomass burning studies, namely: plume-rise, transport, and inverse modeling. Section 4 addresses the need for synergy between satellite measurements and modeling, through the unification of their hitherto disparate parameter systems, in order to make them more compatible and amenable to better comparison and possible integration. Section 5 completes the paper, with a brief